A deep learning method for predicting lead content in oilseed rape leaves using fluorescence hyperspectral imaging

•FHSI was applied to the detection of lead content in oilseed rape leaves.•WT-SDAE model was proposed for deep feature extraction.•The WT-SDAE-SVR model achieved high-precision prediction of Pb concentration. The purpose of this study was to develop a deep learning method involving wavelet transform...

Full description

Saved in:
Bibliographic Details
Published in:Food chemistry Vol. 409; p. 135251
Main Authors: Zhou, Xin, Zhao, Chunjiang, Sun, Jun, Cao, Yan, Yao, Kunshan, Xu, Min
Format: Journal Article
Language:English
Published: England Elsevier Ltd 30.05.2023
Subjects:
Algorithms
Brassica napus
Deep Learning
Fluorescence hyperspectral imaging
Heavy metal lead
Hyperspectral Imaging
Least-Squares Analysis
Nondestructive testing
Oilseed rape
Plant Leaves
Spectroscopy, Near-Infrared - methods
Stacked denoising autoencoder
Wavelet transform
Oilseed rape
Wavelet transform
Heavy metal lead
Stacked denoising autoencoder
Nondestructive testing
Fluorescence hyperspectral imaging
ISSN:0308-8146, 1873-7072, 1873-7072
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•FHSI was applied to the detection of lead content in oilseed rape leaves.•WT-SDAE model was proposed for deep feature extraction.•The WT-SDAE-SVR model achieved high-precision prediction of Pb concentration. The purpose of this study was to develop a deep learning method involving wavelet transform (WT) and stacked denoising autoencoder (SDAE) for extracting deep features of heavy metal lead (Pb) detection of oilseed rape leaves. Firstly, the standard normalized variable (SNV) algorithm was established as the best preprocessing algorithm, and the SNV-treated fluorescence spectral data was used for further data analysis. Then, WT was used to decompose the SNV-treated fluorescence spectra of oilseed rape leaves to obtain the optimal wavelet decomposition layers using different wavelet basis functions, and SDAE was used for deep feature learning under the optimal wavelet decomposition layer. Finally, the best established support vector machine regression (SVR) model prediction set parameters Rp2, RMSEP and RPD were 0.9388, 0.0199 mg/kg and 3.275 using sym7 as the wavelet basis function. The results of this study verified that the huge potential of fluorescence hyperspectral technology combined with deep learning algorithms to detect heavy metals.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0308-8146
1873-7072
1873-7072
DOI:10.1016/j.foodchem.2022.135251